why does desaturated yellow seem darker than saturated yellow if paper can't be brighter than white?

Question

From what I understand, when we apply traditional pigments (i.e., non neon) on paper, we make the paper reflect less light than it was before we applied it.

That is, when we apply a pigment on paper, we are actually putting a filter on paper that absorbs the complement of the color we think we are coloring with, which in almost all cases reduces the amount of light reflected.

This is most obvious when we apply black pigment on paper, since it absorbs lots of light, the paper gets darker and darker as we apply more and more pigment.

What confuses me is the opposite seems to be true for what we call "bright colors", like red, yellow, sky blue, magenta, etc. That is, we think we are making the paper brighter by applying more pigment, which surely can't be the case, since that's like saying we're making a black paper darker while drawing through a camera on negative, where the white is inverted to black.

This is specifically annoying for me for yellow because yellow seems really really bright when it is saturated on paper, whereas paper should be much brighter than yellow because it is white, and pale hues of yellow should be brighter than yellow because they have less blue prevented from hitting your eyes from them. It is really tempting to say that we want a thick layer of yellow to make something brighter, but how does this make any sense?

I mean I have some guesses that if the pigment is applied flat enough, it might behave kind of like a mirror or glossy paper, and reflect light in a more concentrated way rather than scattering it, and make more light hit your eyes than a regular painting surface, which tends to have a tooth to it, but I am not sure how much this contributes to the question.

Can somebody explain why we think we are making the paper brighter by applying what we perceive to be bright colors, but darker when we apply what we perceive to be dark colors, when surely in both cases we are making the paper darker? Or what's more confusing, it seems like we're making it lighter until a certain point, after which it seems like we're getting darker, especially for really strong pigments that look black at full saturation like Phthalo Blue, Prussian Blue, and Phthalo Green.

Answer 1

The human eye is not a perfect measurement instrument. For one as noted before human eye is highly nonlinear. This means that small changes in brightness do not directly correlate with a human sensory input.

Note that humans do not react to all colors in the same way. Making a very long story short: Your eye is optimized for the color yellow.

From what I understand, when we apply traditional pigments(eg non neon) on paper, we make the paper reflect less light than it was before we applied it.

For a CMYK print yes, since its has to be able to mix those colors. This does not apply generally though: If you have a bunch of oils or some other colors, it does not need to apply. So a general yellow pigment can be brighter than your paper.

But as I said yellow is pretty bright because that's what your eye is most sensitive to. Yeah white can be as bright, but your brain can have a hard time making the difference. Now when you scan a physical image then the imaging system may do surprising things, a camera or scanner is by default not an accurate scientific measurement instrument. Although it can be calibrated to be one with some effort.

There are also other factors too, such as what background lighting you have. Are you seeing it in sunlight, and if so where on the globe and what time of day? What color is your paper? Paper for northern Europe is bluer than paper made for countries nearer the equator for example.

Answer 2

One thing: Pigment colors are not transparent, a colored place can reflect more than paper.

That's not probably the case when one measures the total reflected light energy except just in your glossy mirror case. But the apparent lightness can be bigger due the highly nonlinear nature of our color perception.

Your color blending diagram has tints and shades of several common paints. Mechanical mixing of pigment colors is another highly non-linear process, very far from cmyk printing and RGB screens.

I believe we call a color bright or dark due its behaviour. The behaviour isn't a surprising finding which is done after deciding for totally other reasons that an yellow color material is bright and a blue one is dark.

Answer 3

Color is perceived by the reflection of a specific wavelength and frequency of the light spectrum. You don't see what is "absorbed" you see what is reflected.

"Absorb" is just a bad word to use in my opinion. I understand the thought process behind using it, but to be fair light isn't absorbed like a sponge and water. Light either reflects or it stops - nothing "holds" the light as absorption would imply. Probably being a bit pedantic on that though.

To be accurate, when you add black to a piece of white paper, you are covering the areas which reflect resulting in the white areas reflecting light more than the black areas. You are not creating areas which "absorb" light. You are changing the frequency of the light reflection. Similar to if standing in front of a mirror and you paint half of the mirror black... the reflection merely does not happen in the painted area - the light has stopped and isn't being bounced back.

You see yellow because the yellow wavelength of light is reflected off of a surface. The specific yellow will depend upon the wavelength frequency.

Therefore, areas which reflect with more strength will bee seen as more vibrant or brighter. Every part of the light spectrum has an optimum wavelength and frequency for every color - one where a particular color is perceived at a greater degree. The closer you are to that wavelength, the "brighter" the color.

_Wikipedia

Desaturated, dark, or pastel colors simply reflect the specific wavelength for a given color at varying frequencies.

And Vantablack reflects nothing :)